Attributes reduction and rules acquisition in an lattice-valued information system with fuzzy decision

In this paper, we aim to investigate latticevalued information systems with fuzzy decision (LvISFD), where the domain of every condition attribute is a finite lattice. Firstly, we propose the concept of LvISFD by combining dominance relation and lattice structure. Meanwhile, we establish a rough set approach and give a ranking method for all objects in this complex system. Secondly, we address approximation reductions and rules acquisition in LvISFD. Furthermore, an algorithm of the presented reduction approach is constructed. Finally, an illustrative example is given to show the effectiveness of the proposed method, and experiment evaluation is performed by four datasets from UCI. These results of this study will be more valuable to solve practical issues.